1. Field of the Invention
The invention relates to an information processing apparatus for handling many similar files.
2. Description of the Related Art
In recent years with the spread of the so-called GUI environment, applications wherein an end user handles many files(data-file, document file, etc.) are expanding. When an end user handles many files, we face first the problem of “the merge operation of file data of common condition with summed data-file” as mentioned later.
In general a data-summing process is often performed in such a stepwise way that summing is first performed by a segment unit such as a branch, a section, or a year and subsequently all summed data are assembled to acquire a total sum. For instance, usually data summed each product code are first summed each branch and each year, and next the first stage data summed each year and each branch are assembled to perform total sum.
In the data-summing process which is performed in a stepwise way as mentioned above, data which are fixed parameter in the first stage summing are usually excluded from a first stage sum data on the ground that they are shared by all constituent records of the file. For instance we can find cases wherein data the item of which is “branch” or “year” is not included in records of first stage summed data generated each branch and each year.
In the case as described above, when we perform the total data-summing process by assembling all the first stage summed data, we must take measures which incorporate the common data of the first stage summed files into the actual data of the data files which are to be assembled. That is, we must add record fields corresponding to items of the common parameter of the file, and enter the common parameter data in item.
For example, in handling CSV files spreading in recent years, the merging operation of common condition data of first stage summed files, as above described, is executed by the end user operation usually as follows. Explaining an example where each file is generated with “branch” and “year” being the common condition as described above, the merging operation is executed in such a way that new columns the index words of which are “branch” and “year” are inserted in the data file displayed in the form of the table format on a display and subsequently the same data of the common parameter are entered in all cells included in the inserted column. For example in the first stage summed file wherein “Osaka branch” and “2003 year” are the common condition, data “Osaka” and data “2003” are needed to be entered in all cells included in the corresponding columns.
The merging operation of the common condition data in the first stage summed file as described above is a daily job in data processing in general and numbers of the first stage summed files are usually by no means few. Therefore the operations are burdensome or onerous to end users whose primary role is more of various kind of data analysis using data summed up.
Furthermore problems which are troubling together when using many similar files such as the first stage summed files, we face a problem of naming for each file. As many files are to be named, the naming operation is first desired to be as easy or fuss-free as possible. By the reason now for example we used such a method that a definite number of characters are cut out automatically from the beginning of the file so as to be defined as the name. But on the other hand by such a kind of method as mentioned above we cannot identify or discriminate each file from many other similar files.
The primary requirement of the naming for files is that a user can recognize “what” each file is only by its name which is perceived first by a user. That is, the user can recognize directly from its name a combination of characteristic attributes of each file by which we can discriminate it from other files. Now naming for many files by which we can discriminate each file among a group of many similar files brings a great burden to end users as daily work.
Furthermore relating to the problem as described above, as what are troubling when end users handle data files, we face a problem with regard to the method of definition for data attributes such as a unit or a number of an effective digit of numerical data. Data constituting a data file are not complete until numerical data are combined with the attribute data such as unit data. Now when defining a unit for a CSV file, for example, users use a method wherein users only describe it on the file specification or another method wherein users assign a specified raw as the raw on which users describe the unit of each data and subsequently leave the interpretation and definition of the description to a program. But these methods both impose to users a burden which goes far beyond the range of end user operation when retrieving data from data-files and processing them.